Schizophrenia is characterized by abnormalities in perception, working memory, attention and learning. The pathogenesis of the disease involves a distributed neural network including various neurotransmitter systems. Despite the enormous diversity in the pathophysiology of schizophrenia it is possible to isolate specific aspects of altered function that are amenable to study in animal models. One such altered process is the inhibition of the auditory startle response by a smaller preceding auditory pulse or prepulse inhibition (PPI). PPI probes a circuit involving auditory inputs to inferior colicullus and to the caudal pontine reticular formation (PnC), which generates the motor output for the startle response. PPI does not require learning and is used as an operational measure for sensorimotor gating. PPI is strongly modulated by a number of cognitive and limbic mechanisms. Modulatory inputs from medial prefrontal cortex (PFC), mid-line thalamic nuclei, in particular the mediodorsal (MD) nucleus, ventral hippocampus and basolateral (BL) amygdala converge on the nucleus accumbens (Nac). Spiny projection cells in the Nac integrate those inputs and project to the pedunculopontine tegmental nucleus (PPT). PPT is believed to be the final pathway, through cholinergic (muscarinic) inhibitory inputs to PnC, of the modulatory actions on PPI. Animal model studies demonstrate that increased systemic aminergic activity (increased dopamine- DA) in the Nac cause sensorimotor gating failure, similar to that seen in schizophrenic patients. We will explore here two fundamental hypotheses related to the function of the Nac: (i) Even though Nac neurons are not grouped in nuclei, they are grouped functionally according to the structures from which they receive inputs (aim 1) and (ii) hippocampal inputs to Nac neurons gate inputs from PFC. We will then proceed to explore the modulatory action of the ventral subiculum, which is one of the major inputs to the Nac (aim 3). We will perform multisite simultaneous recordings of field potentials and units from cortex, hippocampus, amygdala, Nac, PPT, and PnC, as well as intracellular recordings from the Nac, in anesthetized mice in vivo. Some of the data will be obtained from mice studied in Project III, so that electrophysiological data will have an exact behavioral correlate. This will also help interpret data obtained in Project V. In addition, data from intracellular recordings from Nac neurons is essential to project VI. The rationale for the recording sites and the details of the manipulations is based on the human data obtained in Projects I and II.